2014-09-12 00:09 this sort of critter would be more DocScrutinizer05's domain 2014-09-12 00:09 for me, anything > 12 V is "dangerously high voltage" :) 2014-09-12 00:11 wait until you hear about the 3MHz 12kV part of that power supply ;p 2014-09-12 00:12 oddly, it was easier to figure out, though the feedthroughs are going to leave me broke 2014-09-12 00:13 i think your design has to include tachyons in pi^e-dimensional brane-color space. plenty of them. else, it's obvious you're not serious. 2014-09-12 00:57 cascode ladder 2014-09-12 00:57 1..2kV no problem 2014-09-12 00:58 see wiki cascode 2014-09-12 00:59 https://en.wikipedia.org/wiki/File:Cascode-voltage-ladder.png 2014-09-12 01:01 note the design requirements! e.g. R5 needs to deal with max voltage 2014-09-12 01:02 2kV is easily manageable though 2014-09-12 01:02 whitequark: ^^^ 2014-09-12 01:03 for the DC supply I actually would suggest villard 2014-09-12 01:04 hmmmm, interesting 2014-09-12 01:04 that makes sense 2014-09-12 01:05 a bit complicated though 2014-09-12 01:05 maybe I should go the custom transformer route after all. soak the windings in oil 2014-09-12 01:05 well, your signal you need has some unusual requirements 2014-09-12 01:06 that's true 2014-09-12 01:06 transformer will gove you a zoo of stuff but no sawtooth 2014-09-12 01:06 give* 2014-09-12 01:07 Wattage? aka impedance of load? 2014-09-12 01:08 50pF capacitor essentially 2014-09-12 01:08 lol 2014-09-12 01:08 you get that build on half a eurocard 2014-09-12 01:08 maybe a creditcard size PCB 2014-09-12 01:09 I'd dare to build it on a stamp 2014-09-12 01:09 heh 2014-09-12 01:09 hm 2014-09-12 01:10 high voltage isn't per se a problem 2014-09-12 01:10 should I do just the villard cascade from supply voltage (20V, so, seven stages)? 2014-09-12 01:10 or raise it to 200V 2014-09-12 01:10 and 2kV isn't even high volatge at all 2014-09-12 01:10 or something like that via a transformer first? 2014-09-12 01:10 I'm thinking the latter based on some simluations I did 2014-09-12 01:11 I'd use a boost and then a villard 2014-09-12 01:11 you can get up to 200V from a booster 2014-09-12 01:11 seven stages? 2014-09-12 01:12 20->2000 needs 100x 2014-09-12 01:12 7 * 20V is not 2kV in my book 2014-09-12 01:12 hm 2014-09-12 01:12 doesn't each stage double the voltage? 2014-09-12 01:12 2ooV booster and then a 11 stage villard 2014-09-12 01:12 nope 2014-09-12 01:13 oh, that would explain a few things 2014-09-12 01:13 villard adds Vin for each stage 2014-09-12 01:14 that's why you use same capacitors all the way through ;-) 2014-09-12 01:14 yes, that's what I wondered about 2014-09-12 01:15 gotta finally find something to eat in this shitty village 2014-09-12 01:15 bbl 2014-09-12 01:24 atommann has joined #qi-hardware 2014-09-12 01:26 FDCX has quit [Ping timeout: 264 seconds] 2014-09-12 01:31 or instead of a booster use a "balun" type 1:10 transformer, and two transistors 2014-09-12 01:32 prolly one of the tiniest "active" circuits I know of 2014-09-12 01:36 yes, that's what I was thinking of 2014-09-12 01:37 http://wstaw.org/m/2014/09/12/plasma-desktoplr1987.png 2014-09-12 01:39 could really just use that CCFL transformer, it's like $1 2014-09-12 01:39 and I get to use one villard stage rather than eleven 2014-09-12 01:41 sure. could do that. just doesn't make your life any easier, you need 2kV diodes then ;-P 2014-09-12 01:43 Unit Price USD 0.06840 2014-09-12 01:43 not exactly an issue 2014-09-12 01:43 http://www.digikey.com/product-detail/en/RM2000E-TP/RM2000E-TPMSTR-ND/2686890 2014-09-12 01:44 in bthe end you'll end with somthing that has 5 diodes in series, plus birdseed to make them work when in series, and that will look damn similar to villard 2014-09-12 01:45 then come the capacitors ;-P 2014-09-12 01:47 hmmm right, I wonder how fat should they be 2014-09-12 01:49 1nF@2kV is trivial 2014-09-12 01:58 seems enough for my purpose... 2014-09-12 02:11 what the heck became of diodes and transistors during last 30 some years? 2014-09-12 02:13 anyway SMT and high voltage doesn't go together well. That diode has an isolation gap of <4mm. Sounds scary for a 2kV 2014-09-12 02:17 and ideally you should use a graetz bridge. A flyback diode if you feel lucky and don't worry about efficacy 2014-09-12 02:17 hi 2014-09-12 02:19 also you know that output voltage of those CCFL critters is absolutely not stabilized or warranted in any way? It may be 5 times overvoltage without rated load 2014-09-12 02:19 maybe more 2014-09-12 02:25 FDCX has joined #qi-hardware 2014-09-12 02:28 either you want to stabilize the 2kV DC with a simple secondary->primary feedback, or you dimension your cascode ladder so it can cope with whatever your power supply feeds to it, and you use a feedback from output to control input there 2014-09-12 02:29 I'd suggest the latter since you want it anyway to get a clean signal at output 2014-09-12 02:31 anyway make sure your CCFL booster is at least specified to survive operation without rated load 2014-09-12 02:34 for sec->primary feedback don't fall for the idea I just had for 3 seconds: abusing the cascode basis voltage divider. The cascode introduces massive load to that one, it's not stable 2014-09-12 02:36 you need a dedicated high-Z R-divider sth 10:1 or 20:2 and an according 100..200V Zener 2014-09-12 02:37 you could build a somewhat decent lod for your CCFL booster with that divider as well 2014-09-12 02:37 load* 2014-09-12 02:39 you don't want "floating" outputs with undefined or zero load when it komes to kV 2014-09-12 02:39 comes* 2014-09-12 02:45 ((output voltage of those CCFL critters is absolutely not stabilized)) that's actually by design, they are meant to provide a certain current or a certan power, depending on what you want them for. Neon is constant current, FL is constant power 2014-09-12 04:36 atommann has quit [Ping timeout: 246 seconds] 2014-09-12 04:49 atommann has joined #qi-hardware 2014-09-12 06:09 jluis has joined #qi-hardware 2014-09-12 06:39 wej_ has joined #qi-hardware 2014-09-12 06:39 wej has quit [Ping timeout: 250 seconds] 2014-09-12 07:12 wolfspraul has joined #qi-hardware 2014-09-12 07:20 wej has joined #qi-hardware 2014-09-12 07:21 wej_ has quit [Ping timeout: 250 seconds] 2014-09-12 07:29 rodgort has quit [Quit: ERC Version 5.3 (IRC client for Emacs)] 2014-09-12 07:33 rodgort has joined #qi-hardware 2014-09-12 07:36 wolfspra1l has joined #qi-hardware 2014-09-12 07:37 wolfspraul has quit [Ping timeout: 252 seconds] 2014-09-12 07:40 wolfspra1l has quit [Client Quit] 2014-09-12 07:40 wolfspraul has joined #qi-hardware 2014-09-12 07:49 pcercuei has joined #qi-hardware 2014-09-12 07:52 wolfspraul has quit [Ping timeout: 260 seconds] 2014-09-12 07:53 wolfspraul has joined #qi-hardware 2014-09-12 08:03 wolfspraul has quit [Ping timeout: 245 seconds] 2014-09-12 08:04 wolfspraul has joined #qi-hardware 2014-09-12 08:47 xiangfu has quit [Ping timeout: 260 seconds] 2014-09-12 10:10 atommann has quit [Quit: Leaving] 2014-09-12 11:19 wej has quit [Ping timeout: 250 seconds] 2014-09-12 11:25 wej has joined #qi-hardware 2014-09-12 11:30 wej has quit [Ping timeout: 250 seconds] 2014-09-12 11:34 wej has joined #qi-hardware 2014-09-12 11:41 atommann has joined #qi-hardware 2014-09-12 11:46 woakas1 has quit [Ping timeout: 264 seconds] 2014-09-12 11:52 woakas1 has joined #qi-hardware 2014-09-12 12:30 jluis has quit [Remote host closed the connection] 2014-09-12 12:53 sb0 has quit [Ping timeout: 264 seconds] 2014-09-12 12:58 sb0 has joined #qi-hardware 2014-09-12 13:34 DocScrutinizer05: ah I see, thanks 2014-09-12 13:34 I was going to surround it with an overall feedback loop 2014-09-12 13:35 feedback sawtooth to 220V input of CCFL converter? 2014-09-12 13:35 ;-) 2014-09-12 13:36 feed the primary with voltage-controlled PWM from opamp output. feedback from divider to - input, filtered PWM from controller via optical isolation to + input 2014-09-12 13:37 divide it 1000:1 2014-09-12 13:38 you cannot "dim" those CCFL converters 2014-09-12 13:38 no, not a CCFL converter 2014-09-12 13:38 a CCFL *transformer* 2014-09-12 13:38 that's the same 2014-09-12 13:38 how so? 2014-09-12 13:38 it's just a regular transformer 2014-09-12 13:38 I've seen the DS 2014-09-12 13:39 well, you cannot PWM a transformer, right? 2014-09-12 13:39 I mean it's meant to work with 50Hz sine wave 2014-09-12 13:40 hm 2014-09-12 13:40 I don't know how this mode of operation is called in english 2014-09-12 13:41 you can try to dim it with usual phase-end-enabling triac dimmer 2014-09-12 13:41 ah, what I want to make is essentially a class D amp that feeds the primary 2014-09-12 13:42 but I strongly discourage that 2014-09-12 13:50 what do you think of my class D amp feeding sine to trafo idea? 2014-09-12 13:51 so, then it would have two feedback circuits. one to keep 2kV DC, another for the cascode ladder 2014-09-12 14:02 DocScrutinizer05: anyway I already decided against using CCFL trafo 2014-09-12 14:02 for the reasons you have listed above 2014-09-12 14:02 but it applies just the same to the balun thing 2014-09-12 14:03 I'd use a simple on/of feedback for the Vdd, however you build it 2014-09-12 14:04 hmmm, so feed it with fixed sine, then use a comparator to enable? 2014-09-12 14:04 that makes sense 2014-09-12 14:05 yeah, and a triac/dimmer to only switch in zero crossing 2014-09-12 14:05 if you feed it with AC. Simple on7off for any balun-thing 2014-09-12 14:05 talking about sine, do you think I could use the primary itself as half of LC filter, and feed the filter with rectangular pulses from a FET? 2014-09-12 14:05 I don't have AC 2014-09-12 14:07 when you don't want to use the half-bridge balun-thing, you prolly want to go for a traditional SPSU primary side design 2014-09-12 14:08 using any chip of your choice made for that 2014-09-12 14:09 there are like a million to choose from 2014-09-12 14:10 I'd not build some unique homegrown solution for that 2014-09-12 14:12 we discussed that recently here, with err I forgot who planned to build a power supply for 12V or sth 2014-09-12 14:12 hm, I never used any SPSUs with a trafo 2014-09-12 14:12 only with an inductor 2014-09-12 14:12 add a second coil and you got a trafo 2014-09-12 14:12 well yeah, that's essentially same thing as I want 2014-09-12 14:13 could reuse an existing chip, maybe even with integrated FET 2014-09-12 14:13 and feed it back via a bigger divider 2014-09-12 14:13 yep, exactly 2014-09-12 14:13 for 200V should be simple 2014-09-12 14:13 I think in computer PSUs they optically isolate the secondary from primary? not sure how to do that 2014-09-12 14:14 there are also pretty cheap 12V->220V-AC converters ready made out there 2014-09-12 14:14 I mean, optically isolated feedback 2014-09-12 14:14 optocoupler 2014-09-12 14:14 ah, no, I think it goes the other way 2014-09-12 14:14 they only have FET on primary side and everything else is on secondary side 2014-09-12 14:15 that makes it simpler 2014-09-12 14:16 http://www.reimo.com/de/821275-carbest_inverter_ps_120_12_230v_100w/ 2014-09-12 14:17 100W 2014-09-12 14:17 overkill though 2014-09-12 14:17 I don't even need 1W 2014-09-12 14:17 but supposed to be regulated and low-load-resilient 2014-09-12 14:24 http://www.lelong.com.my/dc-to-ac-power-inverter-module-12v-to-220v-40w-bee-148162984-2015-04-Sale-P.htm#descpContainer 2014-09-12 14:37 ehh, I'll do my own, it doesn't seem very hard 2014-09-12 14:37 also will get SMPS experience :p 2014-09-12 14:38 one more thing to debug, but it's not like I have a deadline 2014-09-12 15:10 fine then, taking pride in DIY is the right spirit ;-) 2014-09-12 15:12 atommann has quit [Quit: Leaving] 2014-09-12 15:37 DocAvalanche has joined #qi-hardware 2014-09-12 15:39 DocScrutinizer51 has quit [Ping timeout: 252 seconds] 2014-09-12 15:41 DocAvalanche has quit [Ping timeout: 252 seconds] 2014-09-12 15:46 DocScrutinizer51 has joined #qi-hardware 2014-09-12 16:10 pcercuei has quit [Ping timeout: 268 seconds] 2014-09-12 16:23 DocScrutinizer05: I just realized one thing 2014-09-12 16:23 you were saying you need to drive a trafo with sine 2014-09-12 16:23 however an SMPS doesn't do that even nearly 2014-09-12 16:25 I said it's supposed to be powered by sine in your particular component 2014-09-12 16:26 when you don't use sine, you get differing results on behavior 2014-09-12 16:26 ah, yes, I understand that 2014-09-12 16:27 but does it really matter if I have feedback anyway? 2014-09-12 16:27 trafos and inductivities have a number of properties and those determine what you may and may not do with them and what are the results you see 2014-09-12 16:27 as in, it will work in a drastically different mode than it is supposed to 2014-09-12 16:27 yes 2014-09-12 16:28 anyway, have to run 2014-09-12 16:28 see you 2014-09-12 16:28 bbl 2014-09-12 16:42 ha! was looking at this trafo: http://www.digikey.com/product-detail/en/760390014/732-4458-1-ND/3831278 2014-09-12 16:43 digikey links right away to this driver: http://www.digikey.com/product-detail/en/SN6501DBVR/296-37700-2-ND/3431175 2014-09-12 16:43 oscillator and two FETs in convenient package 2014-09-12 16:47 hm, 5V though 2014-09-12 16:53 hm, it's completely useless for me 2014-09-12 16:53 but, it gives me ideas 2014-09-12 21:13 wolfspraul has quit [Quit: leaving] 2014-09-12 21:43 zear has quit [Ping timeout: 276 seconds] 2014-09-12 21:44 kyak_ has joined #qi-hardware 2014-09-12 21:44 kyak_ has quit [Changing host] 2014-09-12 21:44 kyak_ has joined #qi-hardware 2014-09-12 21:44 zear has joined #qi-hardware 2014-09-12 21:45 kyak has quit [Ping timeout: 276 seconds] 2014-09-12 22:12 mth_ has joined #qi-hardware 2014-09-12 22:18 sb0_ has joined #qi-hardware 2014-09-12 22:18 sb0 has quit [*.net *.split] 2014-09-12 22:18 FDCX has quit [*.net *.split] 2014-09-12 22:18 qi-bot has quit [*.net *.split] 2014-09-12 22:18 zear has quit [Ping timeout: 260 seconds] 2014-09-12 22:22 zear has joined #qi-hardware 2014-09-12 22:23 valhalla_ has joined #qi-hardware 2014-09-12 22:25 qi-bot has joined #qi-hardware 2014-09-12 22:26 dandon_ has joined #qi-hardware 2014-09-12 22:29 dandon has quit [Ping timeout: 255 seconds] 2014-09-12 22:29 jbenedetto has quit [Ping timeout: 255 seconds] 2014-09-12 22:29 dandon_ is now known as dandon 2014-09-12 22:32 valhalla has quit [*.net *.split] 2014-09-12 22:32 mth has quit [*.net *.split] 2014-09-12 22:38 Luke-Jr has quit [*.net *.split] 2014-09-12 22:39 jbenedetto has joined #qi-hardware 2014-09-12 22:41 dandon_ has joined #qi-hardware 2014-09-12 22:42 wpwrak has quit [Ping timeout: 246 seconds] 2014-09-12 22:42 tumdedum has quit [Ping timeout: 246 seconds] 2014-09-12 22:43 dandon has quit [Ping timeout: 246 seconds] 2014-09-12 22:44 dandon_ is now known as dandon 2014-09-12 22:44 valhalla_ has quit [Ping timeout: 246 seconds] 2014-09-12 22:48 woakas1 has quit [Ping timeout: 276 seconds] 2014-09-12 22:48 Ornotermes has quit [Ping timeout: 276 seconds] 2014-09-12 22:48 sb0_ has quit [Ping timeout: 276 seconds] 2014-09-12 22:48 rozzin has quit [Ping timeout: 276 seconds] 2014-09-12 22:54 wpwrak has joined #qi-hardware 2014-09-12 22:55 Ornotermes has joined #qi-hardware 2014-09-12 23:10 sb0 has joined #qi-hardware 2014-09-12 23:14 tumdedum has joined #qi-hardware 2014-09-12 23:18 jbenedetto has quit [Ping timeout: 252 seconds] 2014-09-12 23:26 jbenedetto has joined #qi-hardware 2014-09-12 23:26 mth_ is now known as mth 2014-09-12 23:38 tumdedum has quit [Changing host] 2014-09-12 23:38 tumdedum has joined #qi-hardware 2014-09-12 23:39 tumdedum has quit [Max SendQ exceeded] 2014-09-12 23:40 tumdedum has joined #qi-hardware 2014-09-12 23:41 Luke-Jr has joined #qi-hardware 2014-09-12 23:43 Luke-Jr is now known as Guest85934 2014-09-12 23:54 woakas1 has joined #qi-hardware 2014-09-12 23:54 valhalla has joined #qi-hardware